Spartan 3A Evaluation Kit

[racerxdl]

Hi all,  I have two Avnet Xilinx Spartan 3A Evaluation Kit here, one (the most used) has now a problem. The U3 (as I see on the schematics the reset controller) is heating TOO much and starts to reset the board constantly. I dont now why, since the current measuring (the test points measured 8mV for 3.3V, so at 0.1R Resistor is 80mA, and 1mV for 1.2V so 10mA) is normal. That happpened from nowhere.

Any ideas what it can be?

My design is running fine at other board with no heat at all at U3.

Thanks!

[bhfletcher]

I have not heard of this issue on any other boards.  U3 is a TI TPS3106K33, which is a voltage supervisor that we use for 1.2V and 3.3V.  Do you see anything unusual on 3.3V or 1.2V?

Bryan

[racerxdl]

I saw now here, its not U3 that is heating, actually the heat on U3 is because the 3.3V Regulator on the other side of the board. What is heating is U5...

But that is a strange thing, happened from nowhere.

The strange thing is that 3.3V voltage is correct, and the voltage at TP1 and TP2 (for current measuring) is half than the other board that I have. I double checked if was anything on the board that was shorting or so, but nothing found (the current is ok, so its not shorting).

And the second strange thing is that PSoC and FPGA boots normally and works correct for about 5 minutes. After that, the two supplies shutdown for 10s, and after that the two suplies start and shutdown with 1s interval between. And doesnt stop until I remove power from the board. I didnt measure the U5 temperature but I can if needed.

[bhfletcher]

U5 is the 3.3V regulator on board, using TI TPS62290.  U5's input power is the 5V USB.  The TPS62290 has a max input voltage of 7V.  Although extremely rare (and well beyond USB spec), I have heard of a couple cases where the TPS62290 may become damaged from a voltage spike on the USB voltage.

It sounds like U5 is damaged.  If there is something funny with the regulation, then the supervisor will trip, wait, and recycle.  I expect replacing U5 will correct the problem.  Sorry for the trouble you are experiencing.

Bryan

[hch]

To add my 2 cents: I awoke my kit from about 2-year-rest today, and after several hours I experienced the same (or very similar) accident.

I forgot to disconnect the board (running the reflashed demo Microblaze code) while browsing the Internet on more projects. I disconnected it when I noticed the blinking LEDs, and later it did not run any more. The kit draws about 370mA from USB (mostly trhu the probably dead IC).

I'm afraid the kit, or its power stage if you prefer, is rather sensitive to power glitches or handling. I really don't look forward to replacing that tiny device with thermal pad at the bottom

H.

[hch]

Just to keep you informed: I changed the U5, and the board is running again. My first experience with such tiny package did not went completely smoothly, but I succeeded in the second try.

I had a backup solution: In case of my inability to replace that ugly device or damage of the PCB I planned to remove the shunt resistor (R13) and mount an old plain 3.3V LDO regulator with caps whenever appropriate.

HTH,

H.